What is it like to roll out a digital service to every country in the world?
Working with the Foreign, Commonwealth and Development Office (FCDO) has been the ultimate exercise in operating at scale. It’s raised some really interesting design and delivery challenges and made me question if I’d ever have another opportunity at a truly global rollout.
In certain countries, if a British national wants to get married abroad they need a document from a UK register office or the British embassy. With an estimated 84,000 UK residents travelling overseas to get married and 12 million British nationals living overseas this can generate a lot of casework and enquiries to the FCDO. I was the product delivery lead on the Caution Your Blast Ltd (CYB) team that designed and built a digital service that improves the guidance to help British nationals understand what documents they need when getting married abroad. The service allows them to apply for a marriage affirmation or certificate of no impediment.
I had a bit of a crisis in confidence when I found out I’d be coordinating a rollout across 226 countries and territories. On a personal level, I’ve lost count of the number of times during this project that I’ve needed to search for where a capital city, country or territory is found. Geography has never been my strength and this project made me reflect on whether there are any other organisations operating at a similar scale as FCDO. I couldn’t think of any. There’s obviously banks and other large globals that spring to mind – but even they don’t have operational knowledge covering every country from Afghanistan to Zimbabwe and all of the territories in between.
We’ve faced some of the obvious challenges that come with global operations – coordination across time zones and the inevitable stretch that creates in working hours to accommodate conversations. Combined with the immediate cost-of delay from missing the crossover window and having to wait a working day to continue a discussion.
We initially split the rollout into regions to help us manage demos and clearances by shifting our working days slightly. This worked well, however some clearances took longer and so, inevitably, we ended up working across multiple regions at once. It meant we needed to be disciplined as a team – prioritising conversations with regions when we had crossover with them.
I haven’t become this familiar with a spreadsheet since the lead-up to my wedding. And I dusted off a lot of the same auto-format and filtering skills that I mastered then. It was critical for each country/territory to be signed off by the relevant embassy or consulate because they understand the varying local marriage policies. The new digital service meant an operational change so we arranged demos to help embassies understand how their ways of working would change. We also needed to manage access to the systems involved for a global network of over 200 people. There were a lot of moving parts alongside designing and iterating the service. We opted for a phased rollout, launching groups of countries as they were cleared. We redirected those countries away from the original guidance and into the new service.
One of the early decisions we made was to create a second testing environment. It meant that staff in embassies could be given access to one environment to check for factual accuracy and we could make rapid, country-specific iterations based on their feedback. We could then do more detailed feature testing ahead of deployments in a consistent pre-production environment.
Time zones and project tracking aren’t unique challenges – what really reinforced the global scale of rollout within FCDO was the variations for each country. At the core of our project was the principle that we could create a single service to allow anyone around the world to apply for the marriage document that they needed. However, there are 3 different documents that might be requested by foreign authorities, plus the scenario that no document is needed. There are then operational variances in who is eligible for those documents and how they are issued by embassies.
The document requested by a foreign authority is outside of FCDO’s immediate control. The operational differences are there for legitimate reasons. In larger countries people can be days away from their closest embassy or consulate. It isn’t practical to expect them to attend an appointment. In other countries, visiting the embassy is more practical than offering a postal option.
We had to do a lot of work on the service flows and logic for routing people to the right journey depending on the country they were getting married in. The result for us was that we needed to build a service that can cope with multiple service combinations based on which document is needed and whether a British national needs to attend in person. Our end solution was to build a service platform delivering five mini services – it triages the people that needed in-person appointments while allowing others to apply online for the correct document.
Building a global service has meant we’ve had FCDO people from around the world reviewing it. Each person brings a different perspective of the common user questions and errors (note: we have also done research with British national users). Previously, each country produced their own guidance on applying for a marriage document in their country.
By significantly broadening our feedback loop through signing-off the global service we’ve been able to apply one country’s content improvements to the global service in a single update. FCDO and their users are now benefiting from an economy of scale with service iterations.
Rolling out to so many countries meant that we had access to an unofficial A/B testing channel. We were able to select a handful of countries to initially launch with and work closely with them to understand whether the live service was behaving as expected – from both an end user and operational perspective. We could gather data and closely monitor if any unexpected challenges arose. Whilst we didn’t encounter anything significant we were able to iterate the internal operational guidance to embassy staff.
As we near the end of this project I feel incredibly lucky to have worked on designing and rolling out the global marriage service. Can I name 226 countries or territories and their capitals? No, absolutely not, I’m not sure I’ve retained any geographical knowledge. But I do know that the teams at CYB can design and deliver at a global scale – spotting the top level service patterns and customising a ‘choose your own adventure’ based on the logic that sits beneath.
Originally posted here
